A Review of Matthijs M. Maas’ Book, Architectures of Global AI Governance

Abstract

This review examines Matthijs M. Maas' work Architectures of Global AI Governance: From Technological Change to Human Choice. The context in which this work is presented is the qualitative leap in the development of AI-based systems since 2016, and especially since 2022, when the pursuit of general-purpose systems intensified amid talk of an arms race. The fundamental question running through the book is how to govern a highly variable technology (where outcomes and risks are extremely context-dependent) that evolves much faster than law and diplomacy. To answer this, Maas constructs a framework of three complementary lenses: the first reveals that regulation fails when it focuses on the artifact rather than on the sociotechnical changes it enables; the second analyzes how that same AI disrupts the legal tools with which we attempt to govern it; and the third describes the actual institutional landscape: not a single, large governing institution, but a polycentric and fragmented ecosystem to which AI governance must adapt and which, in turn, it will reshape. It is this articulation between levels (sociotechnical, legal, and institutional) that serves as the framework from which this review will assess both its strengths and its limitations.